With AI agents going rogue, they can emerge as a risk to the users because they come with abilities to make decisions on their own with minimal human interference. When it comes to the banking industry, they are especially raising critical concerns as some AI agents have reportedly tried to access customer accounts and take financial actions independently. This simply puts your hard earned money at the biggest risks.
Major banks in the United States have started blocking third-party AI agents from accessing customer accounts and carrying out transactions due to potential security risks. According to Reuters, banks have increased concerns that AI agents may put customer’s money and financial information at risk.
Can AI Agents Really Make Decisions About Your Money?
Unlike conventional AI bots, AI agents have far more capabilities to make decisions on your behalf. These may include managing data, access online services, making purchases and even book reservations. However, problems arise when these agents perform actions that users have not instructed them to.
According to an Axios report published on Thursday, some AI agents reportedly tried to take actions that seemed to be unusual and to be independent of users’ instructions. This has raised concerns about what could happen if such technology gains access to sensitive banking information.
For example, if a customer shares their banking login credentials with an AI agent, the technology may be able to access account information and potentially start financial transactions, depending on the permissions and security controls in place. One of the biggest concerns is the security of customer accounts. Some users have reportedly shared their banking usernames and passwords with AI agents to allow them to perform financial tasks.
Reportedly, banks in the US have started to detect and block transactions involving such agents. However, there is another complication. If customers willingly share their login credentials with an AI agent, transactions carried out using those credentials cannot automatically be classified as unauthorised. This creates a difficult situation for banks, as they must determine whether a financial action was genuinely approved by the customer or independently initiated by an AI system.
JPMorgan, Morgan Stanley And Charles Schwab Take Action
Leading US financial institutions have already started to restrict third-party AI agents. Morgan Stanley has blocked outside AI agents from accessing its platforms. Meanwhile, JP MOrgan Chase CEO Jaime Dimon has also warned about the risks of allowing outside AI agents to operate within banking systems. He said, “We’re not going to allow just any outside agent to come in inside the company because that’s just too risky for clients and for JPMorgan.”
Similarly, Charles Schwab CEO Richard Wurster confirmed that the brokerage was blocking unauthorised AI agents from accessing its services.
He said, “I get a report every day tracking how many times we’ve had to block one of the different agents, and it’s not many because I think people aren’t using it in that way in their financial life. We are blocking it because we want to have control over authentication and the experience.”
What Are The Real Risks For Bank Customers?
The growing use of AI agents in financial services could create several risks for customers. One major concern is that an AI agent might misunderstand instructions and attempt to carry out an action the customer never intended.
Another risk involves sharing sensitive banking information. Giving an AI agent access to usernames, passwords or other account details could expose customers to security and privacy problems, particularly if the third-party service is compromised.
There are also concerns about accountability. If an AI agent initiates an unwanted financial transaction using credentials voluntarily provided by a customer, determining who is responsible could become complicated.
For now, US banks are taking preventive measures by restricting unauthorised third-party AI agents. Customers should avoid sharing their banking passwords with AI tools and use only financial services and integrations approved by their banks.
